1. Field Of The Invention
The present invention relates to an apparatus and method for diagnosis of catalyst performance comprising a sensor and an algorithm to process the signals from the sensor. In particular, the present invention relates to a method wherein the performance of an motor vehicle exhaust catalyst is assessed based on a function of the frequency, e.g., the transfer function, related to at least one sensed signal which is based on measurement of a fluctuating parameter having frequency.
2. Description of the Prior Art
Catalysts are commonly used as part of exhaust systems to treat motor vehicle exhaust in order to minimize air pollution. The reduction of pollution from motor vehicles is mandated by the Environmental Protection Agency through the Environmental Protection Act. As part of the process to assure compliance, it is common for various motor vehicle regulatory bodies to mandate tailpipe testing of automobiles on a regular basis. In order to avoid the expense of such emissions, inspection and to assure that automobiles on the road are complying with the environmental laws and regulations, there are efforts to develop a system which can sense when the exhaust system is not compliant and signal the vehicle operator accordingly.
It is mandated to monitor the exhaust gas of a motor vehicle during normal operation to determine whether the catalytic converter is performing as required. The apparatus and method to accomplish this is commonly referred to as on-board diagnostics (OBD). The strategy which is contemplated is that the performance of the catalyst is determined based on sensing the exhaust gases to determine whether the catalyst is performing as specified and required. Different sensing means have been proposed but all are required to signal the motor vehicle operator if the catalyst is failing to operate as required.
Sensors useful to measure various components in gaseous exhaust streams such as motor vehicle exhaust streams are known. Useful sensors include oxygen sensors and NOx sensor assemblies. Such oxygen sensors include on/off and nonlinear sensors known as heated exhaust gas oxygen sensors (HEGO), and universal exhaust gas oxygen sensors (UEGO) which can be used as an on/off sensor or a sensor that provides an output that has a linear relationship to the air to fuel ratio. Various oxygen sensors have been used and are disclosed in the art including the above referenced sensors. Other useful oxygen sensors include a high speed oxygen sensor disclosed in U.S. Pat. No. 5,106,482, a miniaturized oxygen electrode disclosed in U.S. Pat. No. 5,492,611 and lean shift correction of a potentiometric oxygen sensors disclosed in U.S. Pat. No. 5,492,612. U.S. Pat. No. 5,486,336 discloses a nitrogen oxide sensor assembly.
Other references disclosing a variety of sensors include U.S. Pat. Nos. 5,451,371; 5,408,215; 5,444,974; 5,177,464; 5,237,818; 5,452,576; and 4,703,555. These references disclose a variety of sensors including hydrocarbon sensors as well as calorimetric, air to fuel ratio sensors and a general combustion measurement sensor.
Approaches to sense whether the catalyst is performing include the use of dual oxygen sensors. In accordance with this method, one oxygen sensor is located upstream of the catalyst and the other downstream of the catalyst. The signals from the upstream and downstream sensors are compared and correlated to the emissions, typically hydrocarbon emissions, to determine whether the catalyst is functioning to reduce hydrocarbon emissions to achieve compliance with the regulations. If the amount of emissions is calculated to exceed a specific amount, a signal can be sent to the motor vehicle console to alert the operator that the system to treat exhaust has failed and repair is required.
The use of a dual oxygen sensor system has been reported in SAE Technical Paper Series No. 900062, Clemmens, et al., "Detection of Catalyst Performance Loss Using On-Board Diagnostics", presented at the International Congress and Exposition, Detroit, Mich., Feb. 26-Mar. 2, 1990. This paper reviews the history of such systems which are commonly referred to as On-Board Diagnostic Systems (OBD). This early study was indicated to be a proof of concept testing study to identify serious losses in catalyst efficiency with a dual oxygen sensor method. In accordance with this disclosure, testing was conducted at steady state conditions. The results showed that this approach resulted in measurable differences in the pre- and post-oxygen sensor signals between catalyst with good and poor conversion efficiencies.
SAE Technical Paper Series No. 910561, Koupal, et al., "Detection of Catalyst Failure On-Vehicle Using the Dual Oxygen Sensor Method" was presented at the International Congress and Exposition, Detroit, Mich., Feb. 25-Mar. 1, 1991. This paper presents the results of a test program that used a dual oxygen sensor coupled with a simulated On-Board Diagnostic Algorithm to attempt detection of seriously deteriorated catalytic converters on a test vehicle operated over the Federal Test Procedure (FTP). Previous work is reviewed which relates to determining the methodologies to detect catalyst failure by observing the effects of three-way catalyst (TWC) conversion activity on a response pattern generated between an oxygen sensor place upstream and oxygen sensor place downstream of the catalyst. One analytical method is referred to which quantified the fluctuation and sensor response by calculating the area underneath the sensor curves for a discrete time period, then taking the difference in sensor wave form area values. This integrated area difference method has been reported in the above reference to Clemmens. The background further references a study in 1980 by A. H. Meitzler, "Application of Exhaust-Gas-Oxygen Sensors to the Study of Storage Effects in Automotive Three-Way Catalyst", SAE Technical Series No. 800019 which used the response delay of a downstream oxygen sensor to an instantaneous air/fuel shift as an indicator of a catalyst oxygen storage mechanism. Koupal studied the adaptation of the integrated area difference algorithm developed by Clemmens to on-vehicle test results. The results were that the dual oxygen sensor method using integrated area difference analysis, was able to distinguish between good and bad catalyst under controlled conditions.
Presently, the method disclosed in this SAE Technical Paper Series 910561 is a common method of using dual oxygen sensors for on-board diagnostic system measurement. However, it is extremely difficult to obtain consistent in-field measurements due to inconsistencies and, often, insufficient air to fuel swings.
U.S. Pat. No. 5,237,818 is directed to a conversion efficiency measuring apparatus for catalysts used for exhaust gas purification of internal combustion engines. In accordance with the method described therein a reference signal is attained by calculating a correction function of outputs of air to fuel ratio sensors provided at the upstream and downstream ends of the catalyst during an ordinary air to fuel ratio feedback control period. This reference indicates that a drawback to the use of oxygen sensors is that the sensor located downstream of the catalyst is affected by electrical noises, for example, ignition noise, so that this sensor cannot give accurate information of exhaust gases flowing to the sensor.
In Theis, "An Engine Test to Measure the Oxygen Storage Capacity of a Catalyst (provide source), the oxygen storage capacity of a catalyst is assessed to determine its efficacy based on measurements of upstream and downstream air to fuel ratio.
One method to determine whether the catalyst is performing is to measure the switch ratio of the downstream versus the upstream EGO sensors to determine the oxygen storage capacity of a catalytic converter. Based on this measurement the performance can be assessed (J. W. Koupal, M. A. Sabourin, W. B. Clemmens, "Detection of Catalyst Failure On-Vehicle Using the Dual Oxygen Sensor Method", SAE 910561, 1991). Currently, this is the most common method of using dual oxygen sensors for on-board diagnostic system measurements. However, it is extremely difficult to obtain consistent in-field measurements due to inconsistent and often insufficient air to fuel swings.
Another method of using dual oxygen sensors includes biasing the engine air to fuel ratio either rich or lean, and then determining the time it takes for the downstream HEGO sensor to sensor switch in the engine operating condition, vis-a-vis the upstream HEGO sensor. While this is a more reliable method of determining the oxygen storage capacity of the catalytic converter, it is intrusive. The measurement procedure involves changing the operation condition of the vehicle.
There are two inherent problems relating to the use of the oxygen sensors for on-board diagnostic measurements. One, is that there is no strong relationship between the oxygen storage in the catalytic converter and the hydrocarbon conversion performance of the catalytic converter. Secondly, it is difficult to determine a mode of operation of the vehicle under which reliable, meaningful comparisons can be made between the two sensors. It is therefore a continuing goal to devise an algorithm in combination with a sensor strategy to diagnose working of the catalytic converter.
Parent Application U.S. Ser. No. 08/963,171, APPARATUS AND METHOD FOR DIAGNOSIS OF CATALYST PERFORMANCE, filed Nov. 3, 1997 (Docket No. 4077), and which is hereby incorporated by reference is directed to a method and apparatus which assesses the performance of a motor vehicle exhaust catalyst based on the frequency of a signal which is based on measurement of a fluctuating parameter. The method comprises the steps of sensing at least one parameter value as function of time of at least one fluctuating parameter having a frequency in a gaseous or liquid stream. At least one frequency parameter is determined as a function of the at least one parameter value and the frequency. A control value based on the frequency parameter is determined and a control signal based on the control value is generated. Specific frequency parameters identified include the power spectral density and the cumulative power spectral density.